Readers may have become increasingly confused by the claims and counterclaims in relation to the role of blue-light–filtering IOLs and their implications for age-related macular degeneration (AMD). In essence, the dichotomy has produced two schools of thought. The first clearly states that blue-light–filtering IOLs may have profound implications for delaying the processes underlying diseases within the group of AMDs. The second casts doubt on the validity of this concept and expresses further concern regarding the potential consequences of compromising scotopic vision and—perhaps even more importantly—disturbing the diurnal rhythms necessary for ongoing good health. This report hopes to relay the facts in a comprehensive and comprehensible format.
AGING AND AMD
Figures from the World Health Organization (WHO) demonstrate the problems and consequences of a rapidly aging world population. In 2006, one in 10 people worldwide were aged 60 years or older. By 2050, this will rise to one in five people, and by 2150, the figure will be one in three. Consider China, the country with the world's largest population: In 2006, 100 million Chinese were aged 65 years or older, and for the most part they lacked pension provision. In 1958, 300 individuals in the United Kingdom lived to the age of 100 years and received a personal message from the queen congratulating them on their longevity. By 2004, approximately 6,000 individuals reached the age of 100 years, and perhaps the queen considered the use of a word-processor! The projected figures for centenarians in the United Kingdom are 36,000 by 2035 and 1.2 million by 2074.
The consequences of an aging population are particularly apparent when reviewing the prevalence of neuronal diseases. The WHO projects that, by 2025, 100 million people will suffer from senile dementia. In terms of AMD, the projected increase in prevalence is equally disturbing. According to results of the Eurostat study, the prevalence of AMD in the population aged over 50 years is similar for all countries within the European Community (~14%). The predicted figures for 2025, however, show an increased prevalence of AMD between 30% in the United Kingdom to 105% in Ireland. These projected figures indicate mechanisms that are not purely governed by population demographics or genetics.
The implication is that something in our lifestyle is exacerbating the underlying disease process. Large numbers of patients result in high health care costs. In the United Kingdom, the cost that the government ensues for photodynamic therapy (PDT) is approximately £1 million for every 145 treatable AMD patients. Currently, the treatable patients reflect a small fraction of the total number who suffer from AMD. Given the proliferation of sufferers and cost of antiangiogenic pharmaceuticals, AMD could result in health bankruptcy for any governmental system. Hence, there is an increasing interest in AMD prevention.
It is quite clear that AMD is multifactorial. Studies between spouses and siblings—where spouses shared an environment and siblings shared genes—derived the initial evidence that AMD involves genetics. These studies were further supplemented by reports on identical twins. Molecular genetics have now identified 20 or more genes that may be implicated in the disease process. Of particular importance is the number of genes associated with regulatory mechanisms of extracellular matrices of cells (ie, Bruch's membrane). By contrast, epidemiological studies further demonstrate environmental factors that may impact the underlying genetic processes, with particular emphasis on smoking, light, and diet as risk factors. Strong evidence for such environmental risk factors is evident in Japanese studies that showed a huge increase in AMD over 20 years. Clearly, Japanese genes cannot have significantly changed in 20 years, but their environment certainly has.
There is a helpful working hypothesis relating genetic and environmental factors in aging and age-related diseases (Figure 1). In summary, all vital functions at a cellular level are operating maximally at birth. Vital functions decay over a lifetime, with a basal rate determined by each individual's gene control mechanisms for aging. The system may be induced to crash earlier if the individual is exposed to harmful environmental factors. Further, if a late-onset genetic defect is within an individual's gene pool, the system crashes earlier than the basal system—and even earlier in an individual who has accelerated aging by a harmful environmental cofactor (eg, smoking). Clearly, anything that decreases the basal aging rate will delay the onset of symptoms of an age-related defective gene. It is the author's contention that short wavelength radiation is such a harmful risk factor, and its attenuation or avoidance will reduce the rate at which vital functions are lost.
Two discrete mechanisms are seen in cellular aging. In most parts of the body that are exposed to environmental stress (eg, gut lining, surface of the skin), the cell populations are constantly dividing throughout life. Every day, hundreds of thousands of cells are made and lost. In a human lifetime, cells may undergo as many as 12,000 divisions. In a system such as this, aging must be seen as the slow decline in the ability to make new cells. This continues until the system reaches a point where cell death exceeds cell renewal, and the system crashes. The lifetime of vital functions under stress is clearly extended by the constant renewal of cells.
By contrast, in other body parts, cells are formed during embryogenesis and retained by the system throughout life. A prime example of such a system is the brain. The retina is an outgrowth of the brain, and even though it is constantly exposed to optical radiation between 400 nm and 1400 nm, the cells within it cannot divide. As a consequence, aging must be considered as the retention of any genetic errors that enhance the loss of vital functions, together with the accumulated loss of cellular integrity resulting from bombardment by radiation.
THE CELL BIOLOGY OF AMD
Regardless of genetic type, all AMD involves three elements of the outer retina (ie, Bruch's membrane, retinal pigment epithelium[RPE], photoreceptor cells), which are also where changes occur. Subtle changes in the underlying choroidal blood supply are also seen. Any changes in the choroidal blood supply, Bruch's membrane, or RPE may result in compromised transport to the photoreceptor cells and eventually cell death. Pathophysiological studies would suggest the following scenario: light falls onto the photoreceptor cells, is absorbed into the visual pigments in the light-sensitive membranes of the outer segments, and results in the sensation that we know as vision (Figure 2). In the process of absorbing optical radiation in the light-sensitive membranes, however, there is also a production of reactive oxygen species (ROS). These moieties are highly toxic to cellular systems. Given that the photoreceptor cells cannot divide to protect themselves, they have evolved an elegant mechanism to preserve visual function over our lifetime. Every hour of every day, new light-sensitive membranes are made in the junction between the inner and outer segments, and over 2 weeks, such membranes are progressively displaced down the outer segment toward the RPE. On arrival at the RPE, old and potentially radiation-damaged discs are phagocytosed into the pigment epithelial cell. This process seems to be controlled by light—with rods and nighttime receptors being phagocytosed first thing in the morning and cones and photopic receptors being phagocytosed approximately 4 hours into the sleep period. Over a human lifetime, this phagocytic load into what is essentially a nondividing macrophage—the RPE—results in debris build-up.
Because the products of phagocytosis are incompletely broken down, residual bodies known as lipofuscin granules accumulate within the human RPE. These bodies are also light-absorbing systems and, again, a further generation of reactive oxygen species occurs. In turn, the RPE attempts to rid its toxic products through Bruch's membrane. Aging changes within Bruch's membrane, particularly in relation to its collagen content. The capacity of material to pass across Bruch's membrane is reduced, with the net result that an accumulation of debris is deposited within the membrane itself. The depositioning of debris has been shown to significantly restrict the movement of water and micro- and macromolecules across the structure.1 In individuals with genetic defects, focal excrescences of the debris are identified as drusen. In some individuals, this is the initial change in a disease spectrum that may result in geographic atrophy. In others, the disease process goes on to the neovascular form, and more rapid consequences for significant vision loss occur.
The rate of this process may be driven by environmental exposure to light. Van der Shaft et al2 examined the incidence of drusen and disciform lesions in cadaver pseudophakic eyes that received UV-blocking IOLs compared with phakic eyes. This pathophysiological study found a significant increase in both drusen and disciform lesions in the pseudophakic eyes, implying that postcataract surgery exposure to light was a significant risk factor and that UV-blocking IOLs did not offer sufficient filtration.
LIGHT DAMAGE AND AMD
The evidence that light exposure to the retina causes damage is well known; the history extends to the 1960s. Analysis of available data suggests that two types of light damage occur. Type 1 is exposure to low levels of radiation for (1) extremely long periods or (2) intermittent exposures over a long time. The mechanism here seems to be dependent upon absorption of radiation by the visual pigments within the photoreceptor cells. Type 2 exposure is more intense exposure to radiation for shorter periods. It is related to absorption by the lipofuscin within the RPE. Both types of light damage are maximal when the system is exposed to radiation of short wavelengths.
In diurnal animals (ie, animals with rods and cones), the blue cones are the most sensitive and show irreversible damage at levels below other photoreceptor cells. In nocturnal animals (eg, rats and mice), damage occurs as a result of rhodopsin absorption into the rods. In a series of experiments on transgenic mice, Reme et al3 showed that light damage occurred only in animals with rhodopsin in their rods and only after exposure to blue light. No damage with exposure to green light was produced.
Damage to rods and cones, or acceleration of their disc production, will lead to increased phagocytic load in the RPE with increased lipofuscin and propensity to damage from a type 2 mechanism. So, these two systems are intimately interrelated. Furthermore, the type 2 damage described by Bill Ham, and indeed virtually all of the light exposure experimental data, have been collected on healthy young animals.4 In one disturbing report in the Proceedings of the National Academy of Science,5 it was shown that in some inherited retinal diseases, the defective genes make the system even more susceptible to the effects of light. Using typical ophthalmic sources, the authors examined one eye in dogs with a human form of retinitis pigmentosa. These eyes showed a faster degeneration rate than those that were not examined. This suggests that if the mechanism holds true for individuals carrying the AMD gene, they may well be even more susceptible to light damage versus the normal population.
Although no experiment to date has demonstrated the efficacy of blue-light–filtering IOLs in protecting against type 1 damage, there is experimental evidence to show their effectiveness in protecting against type 2 damage. In a series of experiments, Sparrow et al6 exposed human RPE cells fed with an active component of lipofuscin (ie, A2E) to optical radiation either through an AcrySof Natural IOL (Alcon Laboratories, Inc., Fort Worth, Texas) or a UV-blocking IOL. She showed that the AcrySof Natural prevented cell death under such exposure conditions. The counterarguments against this study were that the levels of blue light were more comparable with a retinal exposure from an operating microscope than an exposure from everyday life. This is true, but the mere fact that this system protected from what is potentially a huge overexposure to blue light is an excellent demonstration of the efficacy of such a lens. It would also imply that, given a similar action spectrum between type 1 and type 2, the system would certainly protect against the much lower exposure conditions experienced in type 1 damage.
IS THE ENVIRONMENT SAFE?
The role of epidemiological studies in determining the impact of light damage on AMD has always been a complicated task, given the difficulty in determining the light history of an individual and other confounding cofactors of epidemiological studies. Although many older studies either did not include light exposure or did not conclude that light was a significant risk factor, more modern studies including the Beaver Dam and Blue Mountain group have implicated light as a risk factor. To date, four studies show no correlation, and four studies show a correlation between light exposure and AMD. Thus, clearly, more needs to be done in this area.
When UV-blocking IOLs were introduced, exactly the same counterarguments were employed as those now being levied against blue-light–filtering IOLs, and there is still no direct clinical evidence of the efficacy of UV-blocking IOLs. However, it would be a very brave surgeon that today inserts an IOL without a UV block.
Every ophthalmologist knows that the environment is unsafe if one views the sun directly. We have all seen photoretinitis, often misnamed as a solar burn. Dermatologists have known for some time that the environment is not safe and that daily exposure to solar radiation causes premature skin aging. Indeed, studies have demonstrated a correlation between the incidence of skin cancer and degrees latitude in the United States. As far as ocular exposure is concerned, we must take one further consideration into our argument: Although solar radiation is usually confined to a diurnal rhythm, with 12 hours of light followed by 12 hours of darkness, manmade light sources may be switched on to almost daylight levels at any time of the night or day. In the United Kingdom, in 1990, office lighting levels were set in a standard and remained almost constant until World War II. Following this war and the advent of fluorescent light, lighting levels have increased exponentially and are now commonly 10 or 100 times greater than they were at the end of the 19th century. One only has to look at satellite pictures of the world at night to see that, in most countries, no one is turning the light off any more.
Should we be concerned about extending or interfering with the diurnal aspect of lighting? Well, the lessons from the audiologists are quite clear. Manmade sources of sound have resulted in an increase in hearing loss in the middle-aged population, and this reflects repetitive exposure to a manmade noisy environment during the disco era. Ophthalmologists would do well to appreciate the intensity of the manmade sources that we now use to investigate the fundus. Although most procedures stay below the maximum permissible exposure, some (ie, slit lamp, operating microscope) exceed these levels in approximately 1 minute. This has resulted in microscope manufacturers inserting UV, Infrared, and blue-blocking filters to reduce potential risk to both patients and practitioners. Laser manufacturers also switched from argon blue 488 nm to argon green 514 nm and switched the aiming beams from an attenuated treatment beam to a red beam to avoid any potentially acquired color vision defects in ophthalmologists.7
YELLOW FILTERS IN THE EYE
With increasing age, the natural filtration of the eye is to reduce blue light. The lens, and to a far lesser extent the cornea, show changes that attenuate short wavelength radiation falling upon the retina. By a person's mid-20s, the yellowing of the lens has produced an effective blue blocker. Within the retina itself, there is also a yellow pigment (ie, luteal pigment) that provides a further filter protecting the vital photoreceptor cells of the foveola and fovea. Such is the unimportance of blue light to vision that within the central region of the retina, there are no short wavelength-sensitive cones, and we all effectively suffer from foveal tritanopia. The maximum absorption of the luteal pigment is approximately 450 nm to 460 nm, and therefore, it attenuates in the blue, not the violet or ultraviolet. The aging lens could be described as a consequence of aging rather than a protective mechanism, but it is difficult to argue the same for the luteal pigment, as this is present in all individuals from early childhood. This author argues that there is a real need to attenuate blue light. Indeed, low macular pigment has been identified in many reports as a risk factor in AMD.
DIURNAL RHYTHMS AND SCOTOPIC SENSITIVITY
Much has been made of the body's requirement to respond to the diurnal cycle of the sun by changing circadian rhythms in the body. The recent discovery that 1% of ganglion cells had a photosensitive pigment (ie, melanopsin), with an absorption peak somewhere between 460 nm and 470 nm, suggests that light of this wavelength should not be prevented from reaching the retina. The current range of blue blockers does not attenuate radiation in this waveband. Depending upon the dioptric strength of a blue-light–filtering IOL, it never exceeds that of an individual's natural lens aged in their mid-40s. Given that most cataract surgery is conducted in the over-70-year-old age group, these individuals are going to lose a very significant filter that—if there were any potentially harmful or negative effects—will be enhanced rather than further degraded by the use of an IOL such as the AcrySof Natural. Very few individuals aged in their 40s, 50s, and 60s have sleep disorders, and therefore, arguments to suggest that such lenses would precipitate them are somewhat shallow.
When these IOLs were first introduced, there was a great deal of discussion that they would compromise scotopic vision (ie, vision in starlight, not moonlight). Scotopic vision also requires a pupil in the order of 6.5 mm to 7 mm for maximal efficiency. In a series of studies on contrast sensitivity comparing the pigmented and nonpigmented AcrySof lens, recipients did not perceive the difference. The difference disappeared in the data points of the experiment.8 To date, no individual with a blue-light–filtering IOL has complained of significant visual problems at night or color vision problems during the day.
CONCLUSION
Blue-light–filtering IOLs may offer a means of attenuating the lifetime load of photoreceptor debris presented to the RPE. In doing so, this would reduce the rate of accumulated debris within Bruch's membrane and prolong the lifetime of adequate transport mechanisms sustaining the photoreceptor cells. This may well reduce the aging component of AMD.
Because the cutoff of yellow IOL filters allow the transmission of the peak wavelengths that stimulate melanopsin, to date, there have been no reports of problems of sleep deprivation or ill health due to interference with circadian rhythms. In the future, such reports will be unlikely, given that the natural crystalline lens attenuates far more of the incident radiation and over a broader bandwidth than that of artificial blue-light–filtering IOLs. The argument of interfering with melanopsin absorption is somewhat difficult to understand, given that the accumulation of pigmentation in the natural lens by an individual's late 40s, 50s, and 60s would have a far more striking effect on photohealth if these problems were as important as some investigators have indicated.
It has been very interesting to see reiteration of the arguments from the 1980s, suggesting that any filtration in an IOL may potentially be harmful. Back then, surgeons claimed that the UV filtration elements would interfere with Nd:YAG or argon photocoagulation and may leach out of the lens, thereby poisoning the eye. There were also claims that (1) no direct clinical studies demonstrated efficacy and (2) UV filters should be avoided, given the lack of long-term data. These arguments faded, and today, every IOL has a UV block. In a similar vein, the early arguments concerning potentially compromised scotopic vision with blue-light–filtering IOLs have begun to fade in the face of experimentation, while the lack of direct clinical data and long-term effects will take many years to answer.
Man has certainly evolved in our blue-light environment, given the nature of Raleigh scatter in the sky. Man, however, has no blue perception in the fovea and a life-long progressive loss of blue light (as a result of increasing yellow filtration). Given that the human's light environment has now dramatically changed with the introduction of artificial lights, and given that many of the modern efficient artificial sources (ie, fluorescent batons, halogen lamps, xenon sources) are very rich in blue light, this author would certainly elect to change his natural, yellow-filtering, aging lens for a blue-light–filtering IOL—when the time comes.
John Marshall, PhD, FRCPath, FCOptom (Hon), FRCOphth (Hon), is the Frost Professor of Ophthalmology and Chairman of the Academic Department of Ophthalmology, at St. Thomas' Hospital, and was formerly Sembal Professor of Experimental Ophthalmology at the Institute of Ophthalmology, from 1982 to 1991. Professor Marshall states that he is a paid consultant to Alcon Laboratories, Inc. He may be reached at june.spacey@kcl.ac.uk.
Across the Pond | May 2007
Filter and Don’t Get the Blues
Blue-light–filtering IOLs may help attenuate the lifetime load of photoreceptor debris presented to the retinal pigment epithelium.
John Marshall, PhD, FRCPath, FCOptom (Hon), FRCOphth (Hon)
